Kou et al. SpringerPlus (2016) 5:401 DOI 10.1186/s40064-016-2026-7

RESEARCH Open Access Identification of bacterial communities in sediments of Poyang Lake, the largest freshwater lake in China Wenbo Kou1,2, Jie Zhang1, Xinxin Lu3, Yantian Ma1,2, Xiaozhen Mou3* and Lan Wu1,2*

Abstract play a vital role in various biogeochemical processes in lacustrine sediment ecosystems. This study is among the first to investigate the spatial distribution patterns of bacterial community composition in the sediments of Poy- ang Lake, the largest freshwater lake of China. Sediment samples were collected from the main basins and mouths of major rivers that discharge into the Poyang Lake in May 2011. Quantitative PCR assay and pyrosequencing analysis of 16S rRNA genes showed that the bacteria community abundance and compositions of Poyang Lake sediment varied largely among sampling sites. A total of 25 phyla and 68 bacterial orders were distinguished. Burkholderiales, Gallionel- lales (Beta-), Myxococcales, (Delta-proteobacteria), Sphingobacteriales (Bacteroidetes), Nitrospirales (Nitrospirae), Xanthomonadales (Gamma-proteobacteria) were identified as the major taxa and collectively accounted for over half of annotated sequences. Moreover, correlation analyses suggested that higher loads of total phosphorus and heavy metals (copper, zinc and cadmium) could enhance bacterial abundance in the sediment. Keywords: Bacterial community, Sediment, Poyang Lake, High-throughput sequencing

Background Bacterial community composition (BCC) in freshwater Freshwater lakes are one of the most extensively altered lakes has been extensively investigated, partly because its ecosystems on earth due to changes of climate, hydro- potentials in predicting major biogeochemical functions. logic flow and human activities related processes, such as Early studies have shown that lake sediment BCC may be land-use and nutrient inputs (Carpenter et al. 2011). Lake shaped by physicochemical factors, such as temperature, sediments are important grounds for series of biogeo- stream flow (Bernhard et al. 2005), pH (Lindström et al. chemical transformations of essential nutrients (carbon, 2005) and nutrient concentrations (Bai et al. 2012; Zhang nitrogen and phosphorus) and contaminants (Nealson et al. 2015). In addition, BCC has been found to co-vary 1997; Bouskill et al. 2010). Sediment microorganisms, with metal concentrations in lake sediments (Cummings especially bacteria, play a dominant role in these critical et al. 2003; Bouskill et al. 2010; Sauvain et al. 2014). How- processes. Bacteria-mediated transformations in sedi- ever, the above studies and most available reports were ments lead to active exchange of energy and materials obtained based on investigations of multiple isolated lakes. with the water column and intimately connect sedimen- The relationship between BCC and environmental condi- tary processes with diverse aquatic ecosystem functions tions within individual freshwater lakes, especially those (Ranjard et al. 2000; Urakawa et al. 2000). with large volumes and surface areas has not been fully understood (Yannarell and Triplett 2004; Bouzat et al. 2013). Alternatively, whether environmental factors apply similar impacts on BCC in main lake area and estuarine *Correspondence: [email protected]; [email protected] 1 College of Life Science, Nanchang University, No. 999, Xuefu da Road, zone remain unclear. Hongutang New District, Nanchang 300031, Jiangxi, China Poyang Lake (28°52′21″–29°06′46″N, 116°10′24″– 3 Department of Biological Sciences, Kent State University, No. 800 E. 116°23′50″E), located in northern Jiangxi Province, is the Summit Street, Kent, OH 44240, USA Full list of author information is available at the end of the article largest freshwater lake in China with a storage capacity of

© 2016 Kou et al. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http:// creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Kou et al. SpringerPlus (2016) 5:401 Page 2 of 9

2.95 billion m3 (Fig. 1). The lake covers an area of 4125 km2 chosen to cover both of the main basins of the lake and the and has an average depth of 5.1 m. Poyang Lake is a mouths of major rivers that discharge into Poyang Lake, throughput type of lake, mainly collect freshwater from including two sites from main basin of the lake, i.e., Site tributary rivers, including the Gan, Fu, Xiu, Xin and Rao 1 in Songmenshan Region (29°12′26.9″N, 116°11′28.0″E), Rivers, and discharging into the Yangtze River. In recent and Site 4 in Nanjishan Region (28°55′00.1″N, years, natural and anthropogenic inputs of nutrients and 116°16′44.4″E); four sites from the lake estuaries (mouths xenobiotics have consistently increased. As a result, a of influx rivers), i.e., Site 2 (29°11′27.4″N, 116°00′33.3″E), decreasing gradient of nutrients and heavy metals along Site 3 (28°59′10.6″N, 116°40′10.6″E), Site 5 (28°39′54.4″N, transects from estuaries to main lake basins has estab- 116°9′31.9″E) and Site 6 (28°43′42.2″N, 116°24′34.8″E) lished (Liu et al. 2012; Zhang et al. 2014; Wang and Liang from Xiu, Rao, Fu and Xin River to Poyang Lake, 2015). This may result in spatial variations of sediment respectively. BCCs and their biogeochemical activities, which may in At each sampling site, triplicate surface sediment sam- turn impact the lake ecosystem function. However, so far ples were taken with a grab sampler to obtain a total of 18 no study has been done on sediment BCC in Poyang Lake. samples. Samples were transferred into sterile polyethyl- In this study, 16S rRNA gene-based quantitative PCR ene ziplock bags, put on ice and immediately transported and pyrosequencing were used to examine sediment to laboratory. Large organic debris was removed from the BCCs in Poyang Lake. Our specific goals were to (1) sediments with sterile forceps. Afterwards, samples were investigate horizontal dynamics of BCCs (i.e., relative divided into two aliquots. One aliquot was processed abundance and diversity), (2) examine the potential cor- immediately for measurements of sediment property var- relations between BCCs and environmental factors. iables; and the other aliquot was stored in sterile polypro- pylene tubes at 80 °C for molecular analysis. Methods − Study sites and sample collection Measurement of sediment properties Surface sediments (0–5 cm) were collected in May 2011 Sediment pH was measured on sediment slurry at a 1:2.5 from six sites in Poyang Lake (Fig. 1). These sites were (w/v) sediment: distilled water ratio using a FE20K pH meter (Mettler Toledo) (Rayment and Higginson 1992). For the measurement of organic matter, the dry matter content of sediment was determined after oven-dried at 105 °C for 24 h, then grinded using a mortar and pes- tle and sieved using a 0.25 mm mesh for the following measurements: sediment ash-free-dry-mass (AFDM) was obtained as the subsequent loss of weight after 4 h at 550 °C in a BF51800 muffle furnace (Thermal) (Hesse 1972). Total organic carbon (TOC), total nitrogen (TN) and total phosphorus (TP) contents were analyzed by the Walkley–Black wet oxidation procedure, the microkjel- dahl method and the phosphomolybdic acid blue color method, respectively (Liu et al. 1996). The concentrations of heavy metals including copper (Cu), zinc (Zn), lead (Pb) and cadmium (Cd) were quantified using atomic absorption spectrophotometry after microwave digestion of samples. Briefly, 0.5 g of sieved and dried sediment was added into 9 ml concentrated nitric acid plus 3 ml concentrated hydrochloric acid at 175 °C for 10 min (US EPA 2007). After cooling down, the extracts were centri- fuged at 3000 rpm for 5 min; supernatant was analyzed using an AA800 atomic absorption spectrophotometer (PerkinElmer).

DNA extraction and quantitative PCR Fig. 1 Sampling sites in Poyang Lake. 1 Songmenshan Region, 2 Xiu Microbial genomic DNA was extracted from 0.5 g sedi- River Estuary, 3 Rao River Estuary, 4 Nanjishan Region, 5 Fu River Estu- ment (wet-weight) using a Power Soil DNA extraction ary, 6 Xin River Estuary kit (MoBio) following the manufacturer’s instructions. Kou et al. SpringerPlus (2016) 5:401 Page 3 of 9

The obtained DNA was used as templates for quantita- annotated as chloroplasts and archaea were ignored in tive polymerase chain reaction (qPCR) to determine the further analyses. copy numbers of bacterial 16S rRNA genes. qPCR was Pyrotag sequences were deposited in the NCBI performed with the primer set Eub338F/Eub518R (Fierer Sequence Read Archive (SRA) under the project acces- et al. 2005). Standard curves ranging from 105 to 109 gene sion number SRP033375. copies per μl were obtained by a tenfold serial dilution of linearized plasmids (Takara) containing cloned 16SrRNA Statistical analysis genes that were amplified from Escherichia coli DNA. Physical and chemical variations among sediment sam- R2 value for the standard curves was 0.98, the slope was ples were analyzed using principal components analy- −3.15, which corresponded to an estimated amplifica- sis (PCA) by Canoco 5.0 (Biometrics). Differences of tion efficiency of 104 %. All DNA samples were processed sediment environmental variables and copy numbers of along with negative controls and standards. bacterial 16S rRNA genes among sampling sites were assessed using one-way ANOVA by SPSS 19.0 software Pyrosequencing of bacterial 16S rRNA genes package. The level of statistical significance was p < 0.05. The V4–V6 region of the 16S rRNA gene was PCR Based on taxonomic annotation, sequences were amplified from extracted DNA using universal primers grouped at the order level to construct rarefaction curves 530F and 1100R (Turner et al. 1999). Both primers con- and calculate diversity indices (Mou et al. 2013), includ- tained sequencing adaptor regions; the forward prim- ing Chao1 and Shannon (H′) using the Mothur program ers also contained sample barcodes (Lu et al. 2015). The (Heck et al. 1975). The∫ -Libshuff command in Mothur PCR program included an initial denaturation at 95 °C was used to compare BCCs between sequence libraries. for 3 min, followed by 30 cycles of denaturation at 95 °C To reveal the bacterial distribution patterns among sam- for 30 s, annealing at 58 °C for 1 min and extension at pling sites, a heatmap was generated at the order level by 72 °C for 1 min and a final extension at 72 °C for 5 min. PC-ORD5 (MjM Software), based on the same matrix. PCR amplicons were examined by gel electrophoresis To investigate relationships between sediment BCCs (1 % agarose). Verified amplicons were excised from the and environmental variables, distance based redundancy gels and purified first with a QIAquick gel extraction kit analysis (dbRDA) with Monte Carlo tests was carried out (Qiagen) and then with an Agencourt AMPure XP sys- using the Canoco program for Windows 5.0. Further- tem (Beckman Coulter). Purified PCR amplicons were more, Pearson coefficient correlations between the major quantified using a Quant-iT Picogreen dsDNA Assay kit taxa including copy numbers of bacterial 16S rRNA (Life Technologies). Equimolar amounts of amplicons of genes, Shannon and Chao1 indices and sediment envi- different samples were pooled and pyrosequenced in one ronmental variables were calculated using SPSS 19.0. run using a GS 454 junior sequencing system with unidi- rectional Lib-L chemistry (Roche 454 Life Sciences) (Lu Results et al. 2015). Sediment characteristics Obtained raw sequence reads were processed using Among the ten tested parameters of Poyang lake sedi- the Pipeline Initial Process of the Ribosomal Database ments, six of which have no difference among samples, Project (RDP) to sort and rename sequences based on only AFDM % (ash free dried mass), TP (total phos- sample tags before trimming off the tags and primers phorus), the concentrations of Cu and Cd differentiated from sequences (Cole et al. 2009). Trimmed sequences (ANOVA, p < 0.05) (Additional file 1: Table S1). Principal were processed using the Mothur software package for Components Analysis (PCA) of measured physical chem- quality control and sequence annotations (Schloss et al. ical variables grouped sites 1, 2, 4 and 5 away from sites 3 2009). Briefly, sequences that were shorter than 100 and 6. Generally, sites 3 and 6 had greater concentrations bases or contained ambiguous base calls were excluded of AFDM, TP, Cu and Cd, but shallower water depths for further analysis. Subsequently, chimeric sequences than sites 1, 2, 4 and 5 (Fig. 2; Additional file 1 : Table S1). were removed. After quality control steps, the remain- ing sequence reads were clustered into operational taxo- 16S rRNA gene abundance nomic units (OTUs) at 3 % divergence implemented in Quantitative PCR results showed that bacterial 16S rRNA Mothur. The longest sequence within each OTU group gene abundance varied significantly among six sites was assigned as the representative sequence and blasted (ANOVA, p 0.008). The copy number varied between 10 = 11 against the SILVA SSU database for taxonomic anno- 4.96 × 10 and 4.12 × 10 copies per gram of dry sedi- tation (Pruesse et al. 2007). Since bacterial communi- ment (Fig. 3), with the higher values found for sites 3 and ties were being emphasized in this work, sequences 6, and the lower values for sites 1, 2, 4 and 5. Kou et al. SpringerPlus (2016) 5:401 Page 4 of 9

Recovered OTU taxa were affiliated with 25 phyla, 68 orders and 196 species. Rarefaction analysis of bacterial communities was performed at the order level and all but the site 4 library were approaching plateau (Additional file 2: Fig. S1). Order-level Shannon index (H′) values were similar for all samples and ranged from 2.96 to 3.21 (Table 1, p > 0.05).

Bacterial community structure Over 83 % of the annotated sequences were affiliated with 17 bacterial orders of 6 phyla. Each of these 17 orders accounted for 2 % or more of the total sequences (Fig. 4) and was designated as major taxa. Except for Rhizobi- ales, all orders were found in each sequence library. Bur- kholderiales was the most abundant taxa and accounted for 12.61 % of sequences on average; it is followed by Myxococcales (7.81 %), Sphingobacteriales (7.18 %), Gal- lionellales (6.72 %), Nitrospirales (6.31 %), Xanthomo- nadales (5.60 %) and Desulfuromonadales (5.40 %). Burkholderiales represented the most abundant taxa in sediments of all sites except site 6, where Gallionellales was the most abundant. Desulfuromonadales in the main basin sites (sites 1 and 4; 15.79 and 9.23 %, respectively) Fig. 2 PCA ordination of sediment characteristics were more abundant than those from estuaries (sites 2, 3, 5 and 6, ranged 2.80–4.89 %). Nitrospirales occurred mainly in the sites 5 (11.49 %) and 6 (11.09 %) (Fig. 4). Also, heatmap analysis based on relative abundance of major orders grouped samples into three clades: sites 1–4, sites 2–3, and sites 5–6 (Fig. 4). Furthermore, Lib- shuff analysis based on sequences revealed significant dissimilarities among sequence libraries of the six sites (p < 0.0085, with Bonferroni correction) (Additional file 3: Table S2). Species level annotation was obtained for 3940 sequences (23.5 % of total sequences). Out of the 196 recovered species, Sideroxydans lithotrophicus (10.36 %), Albidiferax ferrireducens (8.86 %), Gallionella capsiferri- formans (7.28 %), Methylobacillus flagellatus (5.13 %) and Nitrosospira multiformis (3.58 %) of Beta-proteobacteria, and bemidjiensis (7.89 %), Anaeromyxobac- ter dehalogenans (7.44 %) and Geobacter lovleyi (3.88 %) Fig. 3 Histogram comparing the copy numbers of bacterial 16S rRNA gene in sediments of Poyang Lake of Delta-proteobacteria were generally found in all sites (Table 2).

Influential factors on bacterial communities Pyrosequencing statistics and alpha‑diversity Pearson correlation analysis revealed that bacterial The average length of 16S rRNA gene pyrotag sequences 16S rRNA gene abundance was significantly corre- (without the primers and adaptors) was 504 bp. A total of lated with several sediment property variables, includ- 19,892 bacterial 16S rRNA gene sequences were obtained ing pH (r = 0.54, p = 0.02), TP (r = 0.70, p = 0.001), Cu and 18,242 remained after quality filtering and chimera (r = 0.69, p = 0.002), Zn (r = 0.50, p = 0.035) and Cd removing. Sequences were grouped into 4394 OTUs (3 % (r = 0.65, p = 0.004) (Table 3). Bacterial diversity (Shan- divergence), with 276–1876 OTUs per sample (Table 1). non index) or richness (Chao1) was not significantly Kou et al. SpringerPlus (2016) 5:401 Page 5 of 9

Table 1 Library coverage estimations and sequence diversity of 16S rRNA genes pyrosequencing

Sampling site Reads OTUs Phylum Order Species H′ Chao1

1 3796 1680 22 54 105 3.21 57.00 2 5547 1326 22 54 81 3.02 55.67 3 6422 1876 24 57 115 3.18 59.63 4 427 276 16 40 40 3.02 45.00 5 1034 556 18 42 44 2.96 51.00 6 1016 572 20 44 41 3.07 59.60 Total 18,242 4394 25 68 196

H′, Shannon index

Fig. 4 Heatmap analysis of bacterial community composition at order level

correlated with any of measured sediment variables Discussion (Table 3). As one of the largest freshwater lakes in China, Poyang To determine the effect of sediment properties on Lake provides a number of important ecological services, BCC, the property variables were analyzed using dbRDA such as flood storage, regulation of the local climate and (Fig. 5), where TN, Pb, Cu and Cd were the most contri- habitats for migratory birds. Moreover, it is extremely bution factors as environmental input. Main basin sites (1 rich in biodiversity (Wu et al. 2011). However, in recent and 4) and tributary sites (2, 3, 5 and 6) were mainly sepa- years, the Poyang Lake has been experiencing the prob- rated along the first dbRDA axis, which explained 38.79 % lems of water quality deteriorating and water-level lower- of fitted variation and correlated with TN and Pb contents ing, largely due to the excessive external nutrient loading of the sediment. In addition, the second axis of dbRDA caused by rapid economic development and agricultural explained 27.20 % of total variation. However, no environ- intensification (Wang et al. 2013). The lake ecosystem is mental factors passed the Monte Carlo significance test. facing the degradation trend. Kou et al. SpringerPlus (2016) 5:401 Page 6 of 9

Table 2 Main bacterial community compositions in the sediments of Poyang Lake at species level

Sample site 1 (%) 2 (%) 3 (%) 4 (%) 5 (%) 6 (%)

Sideroxydans lithotrophicus 3.04 14.68 2.97 8.21 26.67 35.67 Albidiferax ferrireducens 1.45 16.12 4.77 5.97 15.19 2.34 Geobacter bemidjiensis 11.88 5.87 9.07 13.43 3.70 1.75 Anaeromyxobacter dehalogenans 11.74 1.00 10.16 14.18 8.89 14.62 Gallionella capsiferriformans 0.00 19.77 0.23 0 1.85 1.75 Methylotenera mobilis 1.01 4.23 10.09 1.49 1.85 0 Geobacter lovleyi 8.70 1.07 4.22 7.46 2.22 4.68 Nitrosospira multiformis 0.00 9.24 0.94 0 0.00 0 Aquabacterium 4.49 1.79 4.61 2.24 4.07 0 Thiobacillus denitrificans 0.72 0.21 8.60 0.75 0.37 4.09 Arenimonas oryziterrae 0.72 4.58 3.83 0.75 0 0 Geothrix fermentans 2.32 1.43 1.80 0 4.81 0.58 Methylobacillus flagellatus 4.20 0.72 1.33 2.24 4.07 1.17 Desulfuromonas acetoxidans 5.94 0.29 0.70 8.96 0.74 0.58 Sulfuritalea hydrogenivorans 2.75 0.93 2.35 1.49 0 0.58 Opitutus terrae 1.45 2.08 0.86 1.49 2.22 0 Acidovorax avenae 0.29 2.01 0.39 0.75 6.67 0

Table 3 Correlations between major taxa and environmental variables based on Pearson’s product momentum correla- tion coefficient

Variables PH SM AFDM TOC TN TP Depth Cu Zn Pb Cd qPCR 0.537 0.702 0.615 0.761 0.639 0.698 – 0.685 0.500 0.285 0.649 H′ 0.179 0.089 0.306 0.466 0.603 0.560 0.435 0.284 0.517 0.339 0.303 − − Chao1 0.039 0.500 0.209 0.269 0.249 0.766 0.195 0.655 0.536 0.329 0.658 − − Values in italics are different from 0 with a significance level alpha 0.05 = H′, Shannon index

Our study revealed spatial heterogeneity of environ- showed spatial heterogeneity. The 16S rRNA gene copy 11 mental variables in the sediment of Poyang Lake. For numbers was 2.31 × 10 copies per gram of dry sedi- example, contents of TP, Cu and Cd from sites 3 and 6 ment on average in this study (Fig. 3), similar as those of (Rao River and Xin River estuaries) were significantly Taihu Lake, another large and eutrophic lake in China higher than other sites (sites 1 and 4 from main basins (Ye et al. 2009). Samples of sites 3 and 6 (Rao River and of lake, sites 2 and 5 from Xiu River and Fu River estuar- Xin River estuaries) had higher values of 16S rRNA gene ies) (Fig. 2). This finding is similar to a previous report, copy numbers than the other sites; and this pattern was in which the concentrations of TP in sediments collected likely shaped by sediment chemical properties, especially from Rao and Xin River estuaries were higher than those contents of TP, Cu and Cd (Table 3). These findings indi- in the center of the Poyang Lake (Wang and Liang 2015). cated that high loads of organic compounds and heavy The observed high concentrations of TP, Cu and Cd from metals may relate to the increases in bacterial abundance. Rao and Xin River estuaries are consistent with their This is in agreement with the results reported in previous locations, which serve as bases for multiple industrial studies, which showed positive correlations between sed- plants such as copper and phosphate mines (Zhang et al. imentary bacterial abundance and levels of organic mat- 2014). Our results indicated that Rao River and Xin River ter and nutrients (Steger et al. 2011; Zhang et al. 2015). were the main input sources of nutrients and metal pol- In addition, bacterial abundance, also recorded using lutants of Poyang Lake among the 6 sampling sites and quantitative PCR, found the highest bacterial population the together appearance of nutrients and metal pollut- within the most heavy metal polluted sediments (Bouskill ants were simultaneous. et al. 2010). In accordance with variations in sediment conditions, For the spatial distribution of BCCs, heatmap analysis bacterial abundance and community structure also revealed that BCCs of two sites 1 and 4 (Songmenshan Kou et al. SpringerPlus (2016) 5:401 Page 7 of 9

Fig. 5 Distance based redundancy analysis (dbRDA) biplots of the sediment bacterial communities associated with environmental variables. Black and solid circles indicate different samples; grey triangles indicate different taxa

and Nanjishan regions) were similar to each other, sug- Previous studies reported that nitrogen concentra- gesting similar sediment conditions in these two sites tion may have a direct impact on the bacterial compo- (Fig. 4). This result may be explained by their close posi- sition in both lake water column and sediment samples tions (both belong to main basin of Poyang Lake) and the (Haukka et al. 2006; Zhao et al. 2012). Low levels of Pb sufficient mixing of water current throughout the main contamination in anoxic freshwater sediment (Rush Lake basin. However, the BCCs from sites 2 and 3 (Xiu River in USA) may impact the community structure of the cul- and Rao River estuaries) were clustered together in Fig. 4, turable fraction of the indigenous microbes (Grandlic even though their sediment characteristics were signifi- et al. 2006). In addition, BCCs in Lake Geneva were sig- cantly different, which were likely affected by riparian nificantly different between contaminated (High contents inputs or other unknown factors. This result is consist- of Pb, Cu and Cd) and uncontaminated stations, where ent with the idea that local adaptation maybe is favoring sulphate-reducing bacteria and Fe(III)-reducing bacteria particular lineages in specific regions (Bouzat et al. 2013). (Geobacter sp.) were more abundant in the contaminated Furthermore, our data supported that spatial distribution sediments (Haller et al. 2011). In this work, among pH difference of bacterial communities within a lake is due value, certain compounds (AFDM, TOC, TN, TP, Cu, Zn, to shifts in the relative abundance of OTUs rather than Pb and Cr) and water depth, TN and Pb were found to variation in presence/absence of some vital species (Sta- be the most important factors that affected variability of ley et al. 2015). bacterial communities (Fig. 5). However, no significant Kou et al. SpringerPlus (2016) 5:401 Page 8 of 9

correlation was observed between sediment physico- were mainly composed of taxa that are typical to fresh- chemical variables and BCCs. This suggests that BCCs in water sediment, including Burkholderiales, Myxococ- the sediment of Poyang Lake may be synergistically regu- cales, Sphingobacteriales, Gallionellales, Nitrospirales, lated by multiple factors. Xanthomonadales and Desulfuromonadales. Despite variation among sampling locations, in gen- Additional files eral, Burkholderiales (Beta-proteobacteria) was the most abundant taxa in the surface sediments of Poyang Lake. Additional file 1: Table S1 Physicochemical variables of sediments in This finding is consistent with our previous report in Poyang Lake. which Burkholderiales was found to be dominant in the Additional file 2: Fig S1 Rarefaction curves of bacterial pyrosequencing. water column of Poyang Lake (Wu et al. 2012). This taxon Additional file 3: Table S2 Libshuff comparison between the sequences is common in freshwater environments (Newton et al. libraries of samples from six sites. 2011; Staley et al. 2015) and has been found to be able to degrade several aromatic compounds (Pérez-Pantoja et al. 2012). In addition, the observation of abundant Myxo- Authors’ contributions WBK analyzed sequencing data, performed the statistical analysis and drafted coccales, Desulfuromonadales (Delta-proteobacteria), the manuscript. JZ collected the samples and carried out the experiments. Sphingobacteriales (Bacteroidetes), Gallionellales (Beta- XXL performed the 454 sequencing. YTM drafted the manuscript. XZM proteobacteria) and Nitrospirales (Nitrospirae) in the sedi- participated in the design of experiment and helped to draft of manuscript. LW designed the experiment and drafted manuscript. All authors read and ment of Poyang Lake may reflect the metabolic versatility approved the final manuscript. of these groups. Gallionellales was proved to participate in the iron cycle of various waterbody environments Author details 1 College of Life Science, Nanchang University, No. 999, Xuefu da Road, Hon- (Emerson et al. 2010; Krepski et al. 2012). Therefore, gutang New District, Nanchang 300031, Jiangxi, China. 2 Collaborative Innova- the existence of many iron-oxidizing bacteria affiliated tion Center for Poyang Lake Basin Green Development and Water Security, to Gallionellales (Sideroxydans lithotrophicus and Gal- Nanchang University, Nanchang 330031, China. 3 Department of Biological Sciences, Kent State University, No. 800 E. Summit Street, Kent, OH 44240, USA. lionella capsiferriformans, Table 2) demonstrated that the active geological cycle of iron may occur in sediment of Acknowledgements Poyang Lake (Blothe and Roden 2009). Nevertheless, the This work was supported by the National Natural Science Foundation of China (Nos. 31060082, 31260110), Jiangxi Natural Science Foundation of iron content remains to be determined before we are able China (2007GZN1927) and the Kent State University Research Council (to X. to draw this conclusion. The prevalence of Nitrosospira Mou). The authors would like to thank Dr. Z. Y. Kong and other members who multiformis in this study revealed active ammonia-oxidiz- provided their valuable and constructive suggestions in our lab. ing process, while the presence of Nitrospirales members Competing interests have been known as nitrite-oxidizing bacteria, therefore, The authors declare that they have no competing interests. nitrification is intensively involved in the nitrogen cycle of Received: 17 November 2015 Accepted: 17 March 2016 lake sediment (Feng et al. 2013; Shen et al. 2013). Methy- lobacillus flagellates occupied almost all samples in this study, and suggested the methyl utilization and carbon cycling in the sediments (Chistoserdova et al. 2007). The predominant Geobacter bemidjiensis and Geobacter lov- References Bai Y, Shi Q, Wen D, Li Z, Jefferson WA, Feng C, Tang X (2012) Bacterial com- leyi represented acetate or tetrachloroethene (PCE) utili- munities in the sediments of Dianchi Lake, a partitioned eutrophic zation, as well as the reduction of heavy metals (Wagner waterbody in China. PLoS One 7:e37796 et al. 2012; Merkley et al. 2015). Bernhard AE, Colbert D, McManus J, Field KG (2005) Microbial community dynamics based on 16S rRNA gene profiles in a Pacific Northwest estuary and its tributaries. FEMS Microbiol Ecol 52:115–128 Conclusion Blothe M, Roden EE (2009) Composition and activity of an autotrophic Fe(II)- In this study, bacterial abundance and community com- oxidizing, nitrate-reducing enrichment culture. Appl Environ Microbiol 75(21):6937–6940 position displayed large spatial variations in the sedi- Bouskill NJ, Barker-Finkel J, Galloway TS, Handy RD, Ford TE (2010) Temporal ment of Poyang Lake. Bacterial community abundance bacterial diversity associated with metal-contaminated river sediments. was mainly affected by estuaries inputs of both nutrients Ecotoxicology 1:317–328 Bouzat JL, Hoostal MJ, Looft T (2013) Spatial patterns of bacterial community and pollutants, while bacterial community composition composition within Lake Erie sediments. J Great Lakes Res 39:344–351 might be affected by the process of biogeochemical trans- Carpenter SR, Stanley EH, Vander Zanden MJ (2011) State of the world’s fresh- formation. The gradient of nutrients and pollutants, geo- water ecosystems: physical, chemical, and biological changes. Annu Rev Environ Resour 36:75–99 graphically distributed from estuaries inlet to the main Chistoserdova L, Lapidus A, Han C, Goodwin L, Saunders L, Brettin T, Tapia R, basin of Poyang Lake, was formed by estuaries inputs, Gilna P, Lucas S, Richardson PM, Lidstrom ME (2007) Genome of Methy- and afterwards the microbial transformation process fol- lobacillus flagellatus, molecular basis for obligate methylotrophy, and lowed. Sediment bacterial communities in Poyang Lake polyphyletic origin of methylotrophy. J Bacteriol 189(11):4020–4027 Kou et al. SpringerPlus (2016) 5:401 Page 9 of 9

Cole JR, Wang Q, Cardenas E, Fish J, Chai B, Farris RJ et al (2009) The Ribosomal Ranjard L, Poly F, Nazaret S (2000) Monitoring complex bacterial communities Database Project: improved alignments and new tools for rRNA analysis. using culture-independent molecular techniques: application to soil Nucleic Acids Res 37:D141–D145 environment. Res Microbiol 151:1–11 Cummings DE, Snoeyenbos-West OL, Newby DT, Niggemyer AM, Lovley DR, Rayment GE, Higginson FR (1992) Australian laboratory handbook of soil and Achenbach LA, Rosenzweig RF (2003) Diversity of Geobacteraceae species water chemical methods. Inkata Press Pty Ltd, Melbourne, Australia inhabiting metal-polluted freshwater lake sediments ascertained by 16S Sauvain L, Bueche M, Junier T, Masson M, Wunderlin T, Kohler-Milleret R, Diez rDNA analyses. Microb Ecol 46:257–269 EG, Loizeau JL, Tercier-Waeber ML, Junier P (2014) Bacterial communi- Emerson D, Fleming EJ, McBeth JM (2010) Iron-oxidizing bacteria: an environ- ties in trace metal contaminated lake sediments are dominated by mental and genomic perspective. Annu Rev Microbiol 64:561–583 endospore-forming bacteria. Aquat Sci 76:S33–S46 Feng S, Chen C, Wang Q, Zhang X, Yang Z, Xie S (2013) Characterization of Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M et al (2009) Introduc- microbial communities in a granular activated carbon–sand dual media ing mothur: open source, platform-independent, community-supported filter for drinking water treatment. Int J Environ Sci Technol 10:917–922 software for describing and comparing microbial communities. Appl Fierer N, Jackson JA, Vilgalys R, Jackson RB (2005) Assessment of soil microbial Environ Microbiol 75:7537–7541 community structure by use of taxon-specific quantitative PCR assays. Shen T, Stieglmeier M, Dai J, Urich T, Schleper C (2013) Responses of the ter- Appl Environ Microbiol 71:4117–4120 restrial ammonia-oxidizing archaeon Ca. Nitrososphaera viennensis and Grandlic CJ, Geib I, Pilon R, Sandrin TR (2006) Lead pollution in a large, prairie- the ammonia-oxidizing bacterium Nitrosospira multiformis to nitrification pothole lake (Rush Lake, WI, USA): effects on abundance and community inhibitors. FEMS Microbiol Lett 344(2):121–129 structure of indigenous sediment bacteria. Environ Pollut 144:119–126 Staley C, Gould TJ, Wang P, Phillips J, Cotner JB, Sadowsky MJ (2015) Species Haller L, Tonolla M, Zopfi J, Peduzzi R, Wildi W, Poté J (2011) Composition sorting and seasonal dynamics primarily shape bacterial communities in of bacterial and archaeal communities in freshwater sediments with the Upper Mississippi River. Sci Total Environ 505:435–445 different contamination levels (Lake Geneva, Switzerland). Water Res Steger K, Premke K, Gudasz C, Sundh I, Tranvik LJ (2011) Microbial biomass 45:1213–1228 and community composition in boreal lake sediments. Limnol Oceanogr Haukka K, Kolmonen E, Hyder R, Hietala J, Vakkilainen K, Kairesalo T, Haario H, 56:725–733 Sivonen K (2006) Effect of nutrient loading on bacterioplankton commu- Turner S, Pryer KM, Miao VP, Palmer JD (1999) Investigating deep phylogenetic nity composition in lake mesocosms. Microb Ecol 51:137–146 relationships among cyanobacteria and plastids by small subunit rRNA Heck KL, Van Belle G, Simberloff D (1975) Explicit calculation of the rarefaction sequence analysis. J Eukaryot Microbiol 46:327–338 diversity measurement and the determination of sufficient sample size. Urakawa H, Yoshida T, Nishimura M, Ohwada K (2000) Characterization of Ecology 56:1459–1461 depth-related population variation in microbial communities of a coastal Hesse PR (1972) A textbook of soil analysis. Chemical Publishing Company, marine sediment using 16S rDNA-based approaches and quinone profil- New York, NY ing. Environ Microbiol 2:542–554 Krepski ST, Hanson TE, Chan CS (2012) Isolation and characterization of a novel US EPA (2007) Microwave assisted acid digestion of sediments, sludges, soils, biomineral stalk-forming iron-oxidizing bacterium from a circumneutral and oil. http://www.epa.gov/epawaste/hazard/testmethods/sw846/ groundwater seep. Environ Microbiol 14:1671–1680 pdfs/3051a.pdf Lindström ES, Kamst-Van Agterveld MP, Zwart G (2005) Distribution of typical Wagner DD, Hug LA, Hatt JK, Spitzmiller MR, Padilla-Crespo E, Ritalahti KM, freshwater bacterial groups is associated with pH, temperature, and lake Edwards EA, Konstantinidis KT, Loffler FE (2012) Genomic determinants of water retention time. Appl Environ Microbiol 71:8201–8206 organohalide-respiration in Geobacter lovleyi, an unusual member of the Liu G, Jiang N, Zhang L (1996) Soil physical, chemical analysis and description Geobacteraceae. BMC Genom 13:200 of soil profiles. Standards Press of China, Bejjing, pp 33–37 (in Chinese) Wang L, Liang T (2015) Distribution characteristics of phosphorus in the sedi- Liu Q, Hu W, Ge G, Xiong Y, Lai J, Wu L (2012) Contents of nutrients and heavy ments and overlying water of Poyang Lake. PLoS One 10:e0125859 metals in the Poyang Lake during dry season. Resour Environ Yangtze Wang S, Shu J, Ni Z (2013) Investigation on pollution situation and counter- Basin 21:1230–1235 (in Chinese) measures in Poyang Lake. J Environ Eng Technol 3:333–349 (in Chinese) Lu X, Sun S, Zhang Y, Hollibaugh JT, Mou X (2015) Temporal and vertical distri- Wu L, Li M, Guo Y, Yang X (2011) Influence of three gorges project on water butions of bacterioplankton at the Gray’s Reef national marine sanctuary. quality of Poyang Lake. Proc Environ Sci 10:1496–1501 Appl Environ Microbiol 81:910–917 Wu L, Ge G, Zhu G, Gong S, Li S, Wan J (2012) Diversity and composition of the Merkley ED, Wrighton KC, Castelle CJ, Anderson BJ, Wilkins MJ, Shah V, Arbour bacterial community of Poyang Lake (China) as determined by 16S rRNA T, Brown JN, Singer SW, Smith RD, Lipton MS (2015) Changes in protein gene sequence analysis. World J Microb Biotechnol 28:233–244 expression across laboratory and field experiments in Geobacter bemid- Yannarell AC, Triplett EW (2004) Within- and between-lake variability in the jiensis. J Proteome Res 14(3):1361–1375 composition of bacterioplankton communities: investigations using Mou XZ, Jacob J, Lu XX, Robbins S, Sun S, Ortiz JD (2013) Diversity and distribu- multiple spatial scales. Appl Environ Microbiol 70:214–223 tion of free-living and particle-associated bacterioplankton in Sandusky Ye W, Liu X, Lin S, Tan J, Pan J, Li D, Yang H (2009) The vertical distribution of Bay and adjacent waters of Lake Erie Western Basin. J Great Lakes Res bacterial and archaeal communities in the water and sediment of Lake 39:352–357 Taihu. FEMS Microbiol Ecol 70:263–276 Nealson KH (1997) Sediment bacteria: who’s there, what are they doing, and Zhang J, Chen X, Liu Q, Wu L (2014) Distribution and potential risk assessment what’s new? Annu Rev Earth Planet Sci 25:403–434 of heavy metals in main estuaries of Lake Poyang’s tributaries. Resour Newton RJ, Jones SE, Eiler A, McMahon KD, Bertilsson S (2011) A guide to Environ Yangtze Basin 23:95–100 (in Chinese) the natural history of freshwater lake bacteria. Microbiol Mol Biol Rev Zhang J, Yang Y, Zhao L, Xie S, Liu Y (2015) Distribution of sediment bacterial 75:14–49 and archaeal communities in plateau freshwater lakes. Appl Microbiol Pérez-Pantoja D, Donoso R, Agulló L, Córdova M, Seeger M, Pieper DH, Biotechnol 99:3291–3302 González B (2012) Genomic analysis of the potential for aromatic Zhao D, Huang R, Zeng J, Yan W, Wang J, Ma T, Wang M, Wu Q (2012) Diversity compounds biodegradation in Burkholderiales. Environ Microbiol analysis of bacterial community compositions in sediments of urban 14:1091–1117 lakes by terminal restriction fragment length polymorphism (T-RFLP). Pruesse E, Quast C, Knittel K, Fuchs BM, Ludwig W, Peplies J, Glöckner FO World J Microbiol Biotechnol 28:3159–3170 (2007) SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB. Nucleic Acids Res 35:7188–7196